Is Power Apps + Power Automate + Dataverse architecture suitable for large-scale data enrichment?

(0) Share

Report

Posted on by GP-14100838-0

Hi,
We are designing a solution that integrates a daily refreshed Power BI report (backed by a semantic model). This report resembles an Excel table keyed by natural (business) columns, but it does not have a unique row ID. Our aim is to enable users to enrich the report data with their own inputs, producing a combined dataset that is easily exportable.
Proposed Architecture

Frontend: Power Apps
Data Integration: Power Automate
Backend: Dataverse

Intended Workflow

Extract the daily Power BI report via Power Automate.
Store the data in Dataverse.
Generate hash keys from the natural keys—since the source has no unique row IDs—to manage record updates and detect deltas.
Build Power Apps UI with:
- An aggregated (summary) dashboard
- An analytical, row-level data entry view
Enable user-level filtering, so each user sees only the records assigned to their login (email).
Capture and store user enrichment in Dataverse, making the enriched dataset available for export.

Technical Challenges
Power Automate:

Only retrieves ~6,000 records per run, but the report contains over 70,000 rows.
No column for incremental extraction or tracking extraction progress.

Dataverse:

Lacks built-in hash generation, complicating delta detection.
No SQL-style merge or left join for easy upsert operation.
Aggregations are only supported in queries, not as reusable views; must be implemented in the front end.
Import reliability issues: occasional partial loads, limited error transparency.
Table-level backups are not supported; only full environment backups, introducing risk for large/high use datasets.

Power Apps:

Must handle business logic that Dataverse cannot (aggregation, delta management, permissions).
API limits and quotas may be manageable now, but concurrency for 120+ users is a concern.
Operations at 70,000+ row scale may not be reliable.

Main Question
Is the combined use of Power Apps, Power Automate, and Dataverse technically appropriate for this scenario at enterprise scale and this data volume?
We have reviewed general AI recommendations, which endorse this architecture in theory, but detailed implementation concerns (particularly around performance and scale) are less clearly addressed. Has anybody encountered similar use case before?

Categories:

Microsoft Dataverse with Power Apps

I have the same question (0)

All responses (3)

Answers (0)

Sort by

Suggested answer

Michael E. Gernaey 53,963 Moderator on at

Like (0)

Report
Copy link

Link copied!

Hi @GP-14100838-0

Personally I have no issue using a canvas app, but I would not push the data into Dataverse, I would build a Power BI App personally.

I do not know how many rows you will end up with, if its ONLY 70k ok fine, but if you have 100 reports and 70k each then no I would not.

While DV is great for Transactional, it is not really intended to be an Analytical storage unique, hence why we made all the stuff possible to export the data.

So if you want to enrich the data with essentially Transaction data then you need a way to create it either

A) in SharePoint

B) In dataverse but you export from here and connect it to the data you have in power bi which can be even from Snowflake or Fabric

In the end, without reviewing your data, your model the # of rows, users, security of items (row versus column versus table data) roles etc all of which I did for decades at Microsoft and for the Partners etc globally, I would not recommend doing what you are doing.

I would need to truly review it all. Its too easy to just say yes its possible. Id' rather you use Elastic DB, or Fabric with psysql etc.

If these suggestions help resolve your issue, Please consider Marking the answer as such and also maybe a like.

Thank you!
Sincerely, Michael Gernaey

Was this reply helpful? Yes No
Suggested answer

wolenberg_ 1,476 Super User 2026 Season 1 on at

Like (0)

Report
Copy link

Link copied!

Hello there,

Power Apps + Power Automate + Dataverse can support large-scale data enrichment, but for enterprise scenarios with 70,000+ rows and 120+ users, this architecture faces performance and scalability limits. Key challenges include Power Automate’s row limits, Dataverse’s lack of native hash or merge logic, and Power Apps’ delegation constraints. To improve reliability, consider using batch APIs, elastic tables, or offloading heavy operations to Azure SQL or Azure Functions. For long-term scalability, a hybrid architecture may be more suitable.

If you need references you can read this contents:

Optimize performance for bulk operations – Microsoft Learn

Handling large datasets in Power Apps – Xylos

Power Platform Community thread on large-scale enrichment

If this helped or could help others in the community, feel free to give it a like or a kudo — it helps surface useful answers for everyone!

1 people found this reply helpful.

Was this reply helpful? Yes No
Ram Prakash Duraisamy 5,877 Super User 2026 Season 1 on at

Like (0)

Report
Copy link

Link copied!

Hi @GP-14100838-0

PowerApps + Dataverse + PowreAutomate looks good. But for the PowerBI alone i feel its not an good option, because PowerBI have some row restriction i feel around 5000+ because it will throw Threshold error.

Please mark as answer if my suggestion helps.
Subscribe here for More Useful videos : https://www.youtube.com/@rampprakash3991

Was this reply helpful? Yes No